Test Equipment Anonymous (TEA) group therapy thread

[Robert763]

I think it’s very optimistic. The SOA chart suggests that. As always you have to read the entire data sheet.

Aye, 'twas always the way. Once you get above ~10-20% of the rated V_ce or V_ds of a device the SOA limited power dissipation of a device plumets way below the headline figure from the front page of the datasheet, and that's typically at 25ºC T_c before you've derated the device for a realistic operating temperature.

Case in point I was looking at yesterday: FDH055N15A power MOSFET. Datasheet figures, front page: 150 V, 167 A, second page P_D 429W (T_c = 25ºC). Look at the DC SOA at V_ds=30V => I_ds = ~2.1A, therefore P_D = 63W only 15% of the headline power figure at 20% of the headline V_DS and 1.2% of the headline I_DS figure, and that's before derating for operating temperature.

Let's just see those figures side by side:
Claim - 150 V, 167 A, 429W.
Actual - 30V, 2.1A, 63W.

The front page (or two) of a datasheet is basically lies made up by the marketing department.

(And yes, I confess, looking at chunky MOSFETs is a tacit admission that I'm looking at doing a DC Electronic Load design, again.)

Yep, problem is they are optimised for swithing applications. You will do better than that with an IRF520.

This is  good reference https://www.infineon.com/dgdl/Infineon-ApplicationNote_Linear_Mode_Operation_Safe_Operation_Diagram_MOSFETs-AN-v01_00-EN.pdf?fileId=db3a30433e30e4bf013e3646e9381200
IXYS make MOSFET for linear use, the L2 series e.g. IXTH110N10L2 SOA is 30V 20A on a 100V 110A "headline" rated device.

[mnementh]

Tek 568... hmmmm...

Jeebus... produced from '67-87...? I'd guess some govt or aviation specification depended on them...?

Interesting design... mostly just power supplies and a CRT, which was driven directly by the sampling plugin. I see...  you could do ANYTHING you wanted with design of the plugin, and there were no limitations up to the CRT's BW limits. I can see why it would remain relevant in certain circles.

Neat.

mnem
I learned something today.

[Cerebus]

Yep, problem is they are optimised for swithing applications. You will do better than that with an IRF520.

I didn't say that I'd selected it, just that the datasheet had passed in front of my eyes.


Edit: That IXTH110N10L2 is a nice part. I already had some IXYS parts on my forming shortlist but not that one. Better still is the IXTT110N10L2 which is the same in a surface mount pack (I've some Al core pc board which I'm thinking of using to mount the hot bits of this).

[bd139]

Tek 568... hmmmm...

Jeebus... produced from '67-87...? I'd guess some govt or aviation specification depended on them...?

Interesting design... mostly just power supplies and a CRT, which was driven directly by the sampling plugin. I see...  you could do ANYTHING you wanted with design of the plugin, and there were no limitations up to the CRT's BW limits. I can see why it would remain relevant in certain circles.

Neat.

mnem
I learned something today.

Exactly. It’s a Swiss Army knife. Up to 14.5GHz sampling head. No tubes. No ICs. That golden spot in the existence of cool toys.

[capt bullshot]

Discovery was the best of the modern Star Dreck series. At least characters were ambiguous and there wasn't a global reset at the end of each episode.

Fixed that to  a proper German word 

[Specmaster]

Woa  is me, the Microsoft 950 arrived today, in superb condition but sadly upon setting it up and updating the OS it became apparent that it used to be owned by a corporate company who had their own profiles installed to allow them to control certain aspects of the phone. This made very clear during updating the OS with the first update of many it required, when the screen displayed a message that some parameters were being restricted / setup by "my organization"  . A couple of things I noticed after setting up my email client, was that I was unable to click on any links in a email, very useful that. Another is that sometimes sites will freeze and refuse to scroll up or down and I have to crash out.

Dispute has been lodged with the evilbay, more as I get a response from the seller. 

[cyclin_al]

Here's a buttock-clenching "low energy" landing. See if you can guess where the airfield is, then skip through every 30s to see if you got it right

Nope, did not even guess close where the airfield was.  That was a long couple of minutes with the stall warning horn (I assume) going off.

There is no stall warning horn. The nearest is the lack of air noise as you get slow, and that you are using your arm to move the stick rather than your fingertips. Besides, you couldn't rely on such an instrument; I've had all the principal instruments fail in flight, and it is normal to do tuition flights with your instruments covered up. The phrase "flying by the seat of your pants" is literally true

Gliders often fly very close to the stall speed. For many gliders, you go upwards fastest a few knots above stall speed. Now combine that with circling in thermals and that thermals can be unpredictable bubbles, and it isn't difficult to find yourself in a spin. Hence the training for the warning signs plus how to get out of a spin.

The other fun thing in thermals is if they are tight and fast. I've been in one going up at >10kt, banking at 70degrees and pulling 2.5G for several minutes. Plus there was another glider at the same altitude, so to keep separation we looked upwards at the top of his head.

Thanks for adding some detail to that.  Yes, I am familiar with the concept of flying VFR.  However, I lack the experience of doing any of this myself.  No time, too closely related to work, plus my ambitions for the rally car keep me occupied that way.

In that video, what was I hearing then that sounded like a horn or alarm?  I totally understand flying extremely close to stall speed in order to make the airfield.

[cyclin_al]

Canadians or Refugees this looks very tidy? eBay auction: #303887615917 Several other toys are worth a look from the same seller.

Thanks for sharing that.

My initial instinct was to jump all over it.
Second sober thought had me checking the cost of shipping  $285CAD   No thanks 

Is anyone local to the west coast?  That is the only way this would be worth it.

[Cerebus]

Spotted on bidspotter:



Ideal for making your neighbour's teeth rattle.

[cyclin_al]

Here's some examples.

Why homedepot can not be accessed from europe?

Access Denied
You don't have permission to access "http://www.homedepot.com/" on this server.
Reference #18.955e6cc1.1614159297.a9b98ea

I don't know. Guess you'll have to go through a VPN. The other big box home center in the U.S. is Lowes. They have similar items. Try their website. Lowes.com

To get past geofencing, do the Canadian versions work?  The product lines are extremely close to that of the USA, but the prices are in Kanuckistan pesos.
Try  homedepot.ca or rona.ca (Rona is now owned by Lowes).

[cyclin_al]

I have waited a long time for this. While my Belzer engineer's spanners were complete in imperial sizes from 5/32" to 3/4", I was missing some of the metric ones. Until today. 15/16/18/19/20/21/22 aquired in fine condition. Ok, I had a rather warty 15 and a 19 (warty too), they did not fit in and I can bask in the gleaming chrome of completeness soon!

My spanner collection is a motley crew, but it's OK with me anyway, and is fairly complete. What I would need is a set of very thin spanners for working coaxial connectors and such. A bit tempted to simply cut them from appropriate bar stock and heat-treat them. Or one can buy cheap ones and grind to fit. Probably needs a water stone grind to not lose tempering.

Cut from appropriate bar stock ... take a look at Filzer bicycle wrenches.  Metric sizes only though.

EDIT:
@Neomys Sapiens, the Park Tool ones are extremely high quality long-lasting wrenches.  They are not cheap.

I interpreted the request as seeking cheap tools.  I have Filzer and Park Tools.  Park Tools is selected where precision or quality really matters.  Filzer for everything else so I do not go broke.

They are called cone wrenches, since the typical use is for the cone-shaped nuts that fit on a bicycle wheel axle.  The cone shape is the inner bearing race from when individual ball bearings were used in bicycle hubs.

[mnementh]

I think it’s very optimistic. The SOA chart suggests that. As always you have to read the entire data sheet.

Aye, 'twas always the way. Once you get above ~10-20% of the rated V_ce or V_ds of a device the SOA limited power dissipation of a device plumets way below the headline figure from the front page of the datasheet, and that's typically at 25ºC T_c before you've derated the device for a realistic operating temperature.

Case in point I was looking at yesterday: FDH055N15A power MOSFET. Datasheet figures, front page: 150 V, 167 A, second page P_D 429W (T_c = 25ºC). Look at the DC SOA at V_ds=30V => I_ds = ~2.1A, therefore P_D = 63W only 15% of the headline power figure at 20% of the headline V_DS and 1.2% of the headline I_DS figure, and that's before derating for operating temperature.

Let's just see those figures side by side:
Claim - 150 V, 167 A, 429W.
Actual - 30V, 2.1A, 63W.

The front page (or two) of a datasheet is basically lies made up by the marketing department.

(And yes, I confess, looking at chunky MOSFETs is a tacit admission that I'm looking at doing a DC Electronic Load design, again.)

Yep, problem is they are optimised for switching applications. You will do better than that with an IRF520.

This is  good reference https://www.infineon.com/dgdl/Infineon-ApplicationNote_Linear_Mode_Operation_Safe_Operation_Diagram_MOSFETs-AN-v01_00-EN.pdf?fileId=db3a30433e30e4bf013e3646e9381200
IXYS make MOSFET for linear use, the L2 series e.g.  SOA is 30V 20A on a 100V 110A "headline" rated device.

It appears the cheap & cheerful toy load (oscillation from driver issue notwithstanding) that started the discussion uses a IRFP260N; this device doesn't even have a DC bracket in the SOA. 

Obvious guess is chosen because they had a cheap & ready supply of either counterfeit or reclaimed parts.

Another end-user subbed a FDH44N50 and claims successfully able to dump 275-300W sustained; it has a much more usable DC SOA (than what I'm guessing the IRFP260N is good for looking at the switch-mode SOA), but it appears that is limited by how much we can drive it using the existing circuit.

It has considerable overhead both in current and voltage directions at ~15V/20A where I want to operate for the most part. Linear derating is 2W/°C on top of the usual 5W/°C so really need to be aggro with the cooling. Maybe hack a old mini-fridge for chilled water and... 

As it's a $7 part vs $19 for the IXTH110N10L2, I feel a lot better aboot grabbing a few of that to slap in parallel for my bass-ackwards DC load.

mnem
 

[bd139]

Woa  is me, the Microsoft 950 arrived today, in superb condition but sadly upon setting it up and updating the OS it became apparent that it used to be owned by a corporate company who had their own profiles installed to allow them to control certain aspects of the phone. This made very clear during updating the OS with the first update of many it required, when the screen displayed a message that some parameters were being restricted / setup by "my organization"  . A couple of things I noticed after setting up my email client, was that I was unable to click on any links in a email, very useful that. Another is that sometimes sites will freeze and refuse to scroll up or down and I have to crash out.

Dispute has been lodged with the evilbay, more as I get a response from the seller. 

Standard windows phone then . Sounds like it has an MDM profile on it at which point it's a brick.

[cyclin_al]

I have waited a long time for this. While my Belzer engineer's spanners were complete in imperial sizes from 5/32" to 3/4", I was missing some of the metric ones. Until today. 15/16/18/19/20/21/22 aquired in fine condition. Ok, I had a rather warty 15 and a 19 (warty too), they did not fit in and I can bask in the gleaming chrome of completeness soon!

My spanner collection is a motley crew, but it's OK with me anyway, and is fairly complete. What I would need is a set of very thin spanners for working coaxial connectors and such. A bit tempted to simply cut them from appropriate bar stock and heat-treat them. Or one can buy cheap ones and grind to fit. Probably needs a water stone grind to not lose tempering.

I have done this more times than I can count. Needs a 8-12" bench grinder, a coffee-can full of water, and patience. Grind with hard pressure for short periods; edges starting to turn black is okay, glowing is not. Quench frequently. Do minimal deburring with using diamond/emery drums on a Dremel; leaving the edges sharp invites the chrome-plating on unground edges to flake off and become little knife blades.

Black Hawk, New Britain, Husky and GearWrench are favorite sacrificial brands for this indignity.

mnem
Stay hungry; stay toolish.

On the topic of deburring tools, would you recommend the same thing for copper tubing used in plumbing, or aluminum alloys?
I tried to use my el-cheapo hand deburring tool on a recent hot water tank replacement job, and it was useless.  I may have previously abused it deburring bicycle handlebars after trimming to length...

[mnementh]

Canadians or Refugees this looks very tidy? eBay auction: #303887615917 Several other toys are worth a look from the same seller.

Thanks for sharing that. My initial instinct was to jump all over it. Second sober thought had me checking the cost of shipping  $285CAD

  No thanks  Is anyone local to the west coast?  That is the only way this would be worth it.

         ROAD TRIP!!!   

mnem
we all thought it. I just said it.

[tautech]

I have waited a long time for this. While my Belzer engineer's spanners were complete in imperial sizes from 5/32" to 3/4", I was missing some of the metric ones. Until today. 15/16/18/19/20/21/22 aquired in fine condition. Ok, I had a rather warty 15 and a 19 (warty too), they did not fit in and I can bask in the gleaming chrome of completeness soon!

My spanner collection is a motley crew, but it's OK with me anyway, and is fairly complete. What I would need is a set of very thin spanners for working coaxial connectors and such. A bit tempted to simply cut them from appropriate bar stock and heat-treat them. Or one can buy cheap ones and grind to fit. Probably needs a water stone grind to not lose tempering.

Cut from appropriate bar stock ... take a look at Filzer bicycle wrenches.  Metric sizes only though.

EDIT:
@Neomys Sapiens, the Park Tool ones are extremely high quality long-lasting wrenches.  They are not cheap.

I interpreted the request as seeking cheap tools.  I have Filzer and Park Tools.  Park Tools is selected where precision or quality really matters.  Filzer for everything else so I do not go broke.

They are called cone wrenches, since the typical use is for the cone-shaped nuts that fit on a bicycle wheel axle.  The cone shape is the inner bearing race from when individual ball bearings were used in bicycle hubs.

No new fangled bike wrenches here only the same one I had as a kid and it was old then ! 

[cyclin_al]

Canadians or Refugees this looks very tidy? eBay auction: #303887615917 Several other toys are worth a look from the same seller.

Thanks for sharing that. My initial instinct was to jump all over it. Second sober thought had me checking the cost of shipping  $285CAD

  No thanks  Is anyone local to the west coast?  That is the only way this would be worth it.

         ROAD TRIP!!!   

mnem
we all thought it. I just said it.

I thought about it for a millisecond.  I can get to Winnipeg Winterpeg in 2.5 days in the car.  So, I am guessing 5 days one way, or 10 days round trip.
That would be ideal if it was summer vacation and there were no health concerns with travelling; that would make a family vacation to see the country.

Dare I ask SWMBO about camping overnights in a tent .... during winter?  Final decision:  NO    (I would be the horse in the emoji)

[Cerebus]

On the topic of deburring tools, would you recommend the same thing for copper tubing used in plumbing, or aluminum alloys?
I tried to use my el-cheapo hand deburring tool on a recent hot water tank replacement job, and it was useless.  I may have previously abused it deburring bicycle handlebars after trimming to length...

Are you talking about one of the Nobo style tools like this?



They will work fine on copper pipe, BUT the softer the material the more practice you need to do a good job with them. I've always used a bit of steel wool or Scotchbrite for copper pipe.

[cyclin_al]

I have waited a long time for this. While my Belzer engineer's spanners were complete in imperial sizes from 5/32" to 3/4", I was missing some of the metric ones. Until today. 15/16/18/19/20/21/22 aquired in fine condition. Ok, I had a rather warty 15 and a 19 (warty too), they did not fit in and I can bask in the gleaming chrome of completeness soon!

My spanner collection is a motley crew, but it's OK with me anyway, and is fairly complete. What I would need is a set of very thin spanners for working coaxial connectors and such. A bit tempted to simply cut them from appropriate bar stock and heat-treat them. Or one can buy cheap ones and grind to fit. Probably needs a water stone grind to not lose tempering.

Cut from appropriate bar stock ... take a look at Filzer bicycle wrenches.  Metric sizes only though.

EDIT:
@Neomys Sapiens, the Park Tool ones are extremely high quality long-lasting wrenches.  They are not cheap.

I interpreted the request as seeking cheap tools.  I have Filzer and Park Tools.  Park Tools is selected where precision or quality really matters.  Filzer for everything else so I do not go broke.

They are called cone wrenches, since the typical use is for the cone-shaped nuts that fit on a bicycle wheel axle.  The cone shape is the inner bearing race from when individual ball bearings were used in bicycle hubs.

No new fangled bike wrenches here only the same one I had as a kid and it was old then ! 

What do you mean new fangled?  I fail to see any real difference between what you have and a new one. 

Where's its mate?  They were always used in pairs, with slightly different sizes.

Front hubs would use 13mm on the cone and 15mm on the locknut.
Rear hubs would use 15mm on the cone and 17mm on the locknut.
The small ones would be 8mm or 10mm for brakes/shifters/cables.

[Saskia]

I think it’s very optimistic. The SOA chart suggests that. As always you have to read the entire data sheet.

Aye, 'twas always the way. Once you get above ~10-20% of the rated V_ce or V_ds of a device the SOA limited power dissipation of a device plumets way below the headline figure from the front page of the datasheet, and that's typically at 25ºC T_c before you've derated the device for a realistic operating temperature.

Case in point I was looking at yesterday: FDH055N15A power MOSFET. Datasheet figures, front page: 150 V, 167 A, second page P_D 429W (T_c = 25ºC). Look at the DC SOA at V_ds=30V => I_ds = ~2.1A, therefore P_D = 63W only 15% of the headline power figure at 20% of the headline V_DS and 1.2% of the headline I_DS figure, and that's before derating for operating temperature.

Let's just see those figures side by side:
Claim - 150 V, 167 A, 429W.
Actual - 30V, 2.1A, 63W.

The front page (or two) of a datasheet is basically lies made up by the marketing department.

(And yes, I confess, looking at chunky MOSFETs is a tacit admission that I'm looking at doing a DC Electronic Load design, again.)

Yep, problem is they are optimised for switching applications. You will do better than that with an IRF520.

This is  good reference https://www.infineon.com/dgdl/Infineon-ApplicationNote_Linear_Mode_Operation_Safe_Operation_Diagram_MOSFETs-AN-v01_00-EN.pdf?fileId=db3a30433e30e4bf013e3646e9381200
IXYS make MOSFET for linear use, the L2 series e.g.  SOA is 30V 20A on a 100V 110A "headline" rated device.

It appears the cheap & cheerful toy load (oscillation from driver issue notwithstanding) that started the discussion uses a IRFP260N; this device doesn't even have a DC bracket in the SOA. 

Obvious guess is chosen because they had a cheap & ready supply of either counterfeit or reclaimed parts.

Another end-user subbed a FDH44N50 and claims successfully able to dump 275-300W sustained; it has a much more usable DC SOA (than what I'm guessing the IRFP260N is good for looking at the switch-mode SOA), but it appears that is limited by how much we can drive it using the existing circuit.

It has considerable overhead both in current and voltage directions at ~15V/20A where I want to operate for the most part. Linear derating is 2W/°C on top of the usual 5W/°C so really need to be aggro with the cooling. Maybe hack a old mini-fridge for chilled water and... 

As it's a $7 part vs $19 for the IXTH110N10L2, I feel a lot better aboot grabbing a few of that to slap in parallel for my bass-ackwards DC load.

mnem
 

Can you give me a list of suitable mosfets ? I have some types floating around and may be able to help out, but currently no time to read any data sheets

[tautech]

I have waited a long time for this. While my Belzer engineer's spanners were complete in imperial sizes from 5/32" to 3/4", I was missing some of the metric ones. Until today. 15/16/18/19/20/21/22 aquired in fine condition. Ok, I had a rather warty 15 and a 19 (warty too), they did not fit in and I can bask in the gleaming chrome of completeness soon!

My spanner collection is a motley crew, but it's OK with me anyway, and is fairly complete. What I would need is a set of very thin spanners for working coaxial connectors and such. A bit tempted to simply cut them from appropriate bar stock and heat-treat them. Or one can buy cheap ones and grind to fit. Probably needs a water stone grind to not lose tempering.

Cut from appropriate bar stock ... take a look at Filzer bicycle wrenches.  Metric sizes only though.

EDIT:
@Neomys Sapiens, the Park Tool ones are extremely high quality long-lasting wrenches.  They are not cheap.

I interpreted the request as seeking cheap tools.  I have Filzer and Park Tools.  Park Tools is selected where precision or quality really matters.  Filzer for everything else so I do not go broke.

They are called cone wrenches, since the typical use is for the cone-shaped nuts that fit on a bicycle wheel axle.  The cone shape is the inner bearing race from when individual ball bearings were used in bicycle hubs.

No new fangled bike wrenches here only the same one I had as a kid and it was old then ! 

What do you mean new fangled?  I fail to see any real difference between what you have and a new one. 

Where's its mate?  They were always used in pairs, with slightly different sizes.

Front hubs would use 13mm on the cone and 15mm on the locknut.
Rear hubs would use 15mm on the cone and 17mm on the locknut.
The small ones would be 8mm or 10mm for brakes/shifters/cables.

Metric ? 
These things were stamped out way before we knew what metric was !
Good ol' Whitworth sizes.

[mnementh]

I have waited a long time for this. While my Belzer engineer's spanners were complete in imperial sizes from 5/32" to 3/4", I was missing some of the metric ones. Until today. 15/16/18/19/20/21/22 aquired in fine condition. Ok, I had a rather warty 15 and a 19 (warty too), they did not fit in and I can bask in the gleaming chrome of completeness soon!

My spanner collection is a motley crew, but it's OK with me anyway, and is fairly complete. What I would need is a set of very thin spanners for working coaxial connectors and such. A bit tempted to simply cut them from appropriate bar stock and heat-treat them. Or one can buy cheap ones and grind to fit. Probably needs a water stone grind to not lose tempering.

I have done this more times than I can count. Needs a 8-12" bench grinder, a coffee-can full of water, and patience. Grind with hard pressure for short periods; edges starting to turn black is okay, glowing is not. Quench frequently. Do minimal deburring with using diamond/emery drums on a Dremel; leaving the edges sharp invites the chrome-plating on unground edges to flake off and become little knife blades.

Black Hawk, New Britain, Husky and GearWrench are favorite sacrificial brands for this indignity.

mnem
Stay hungry; stay toolish.

On the topic of deburring tools, would you recommend the same thing for copper tubing used in plumbing, or aluminum alloys?
I tried to use my el-cheapo hand deburring tool on a recent hot water tank replacement job, and it was useless.  I may have previously abused it deburring bicycle handlebars after trimming to length...

LOL... I'm a heathen when it comes to deburring. I recommended the grinder because I know anything less will fail miserably with the hard heavy chrome-plating which is typical of most decent hand wrenches.

 

https://www.amazon.com/Mundial-SC0500-3-Paring-Knife-Collection/dp/B000ITVW8M

https://www.amazon.com/Mundial-SC0547-4-4-Inch-Paring-Collection/dp/B003AQAZME

Anything soft enough, I typically use one of the cheap restaurant-supply resin-handled paring knives I keep on my bench; it's usually a lot easier to keep the shavings under control. My current one is made by Mundial. They have just the right mix of hard enough to cut most soft metals while still tough enough to not chip out constantly.

   

https://www.amazon.ca/Countersink-Titanium-Five-Blade-Woodworking-Muye/dp/B086HNR7T9

Other tools I commonly use for deburring are needle files and the cheap titanium-nitrided countersinks. They're readily available and you don't cry if you ding one by accidentally trying to deburr something too hard.



https://www.amazon.com/dp/B082TMDSHM

And these work amazingly well on anything small enough to fit inside the throat. Even wheel studs.

mnem

[tautech]

Got something like one of these triangular jobbies and it works pretty good:

[SkyMaster]

Canadians or Refugees......

So says the convict ! 

So the World knows how badly we control our border in Canada... 

The situation is shameful 


 

[mnementh]

I think it’s very optimistic. The SOA chart suggests that. As always you have to read the entire data sheet.

Aye, 'twas always the way. Once you get above ~10-20% of the rated V_ce or V_ds of a device the SOA limited power dissipation of a device plumets way below the headline figure from the front page of the datasheet, and that's typically at 25ºC T_c before you've derated the device for a realistic operating temperature.

Case in point I was looking at yesterday: FDH055N15A power MOSFET. Datasheet figures, front page: 150 V, 167 A, second page P_D 429W (T_c = 25ºC). Look at the DC SOA at V_ds=30V => I_ds = ~2.1A, therefore P_D = 63W only 15% of the headline power figure at 20% of the headline V_DS and 1.2% of the headline I_DS figure, and that's before derating for operating temperature.

Let's just see those figures side by side:
Claim - 150 V, 167 A, 429W.
Actual - 30V, 2.1A, 63W.

The front page (or two) of a datasheet is basically lies made up by the marketing department.

(And yes, I confess, looking at chunky MOSFETs is a tacit admission that I'm looking at doing a DC Electronic Load design, again.)

Yep, problem is they are optimised for switching applications. You will do better than that with an IRF520.

This is  good reference https://www.infineon.com/dgdl/Infineon-ApplicationNote_Linear_Mode_Operation_Safe_Operation_Diagram_MOSFETs-AN-v01_00-EN.pdf?fileId=db3a30433e30e4bf013e3646e9381200
IXYS make MOSFET for linear use, the L2 series e.g.  SOA is 30V 20A on a 100V 110A "headline" rated device.

It appears the cheap & cheerful toy load (oscillation from driver issue notwithstanding) that started the discussion uses a IRFP260N; this device doesn't even have a DC bracket in the SOA. 

Obvious guess is chosen because they had a cheap & ready supply of either counterfeit or reclaimed parts.

Another end-user subbed a FDH44N50 and claims successfully able to dump 275-300W sustained; it has a much more usable DC SOA (than what I'm guessing the IRFP260N is good for looking at the switch-mode SOA), but it appears that is limited by how much we can drive it using the existing circuit.

It has considerable overhead both in current and voltage directions at ~15V/20A where I want to operate for the most part. Linear derating is 2W/°C on top of the usual 5W/°C so really need to be aggro with the cooling. Maybe hack a old mini-fridge for chilled water and... 

As it's a $7 part vs $19 for the IXTH110N10L2, I feel a lot better aboot grabbing a few of that to slap in parallel for my bass-ackwards DC load.

mnem
 

Can you give me a list of suitable mosfets ? I have some types floating around and may be able to help out, but currently no time to read any data sheets

In all honesty... I'm just guessing for this application myself; I've never done anything that sinks this kind of power with a MOSFET. I'm going to build on what has already been resolved and go from there.

My plan was to do the gate resistor and zener mod suggested in the Cheezy DC Load thread, then parallel the two IRFP260Ns that came with on one of them, along with the same mod using a couple FDH44N50s on the other. See what comes of it, and see if it is possible to push either voltage or current appreciably from my intended operating parameters.

That is as far as I've gotten. I'm still shopping around for a decent deal on the FDH44N50s; I haven't even done a parametric search on DigiKey, which would be the obvious way to generate such a list to choose from.

Cheers,

mnem